1. Field of the Invention
The present invention relates to a solid-state imaging device, a method for manufacturing the solid-state imaging device, and an imaging apparatus.
2. Description of the Related Art
Along with an increase in the number of pixels of a solid-state imaging device (image sensor), a size reduction of a pixel has been developed.
On the other hand, an improvement of moving image characteristics through high-speed imaging has been developed at the same time.
In the case where the pixel is miniaturized and high-speed imaging is conducted, the number of photons incident on one pixel decreases and desensitization occurs.
Furthermore, regarding a surveillance camera, there is a demand for a camera suitable for photographing in a dark place. Here, a high-sensitivity sensor is desired.
One example of attempts to enhance sensitivity is signal amplification through avalanche multiplication.
For example, there is an attempt to effect avalanche multiplication of photons through application of a high voltage (refer to IEEE Transactions Electron Devices Vol. 44, NO. 10, October, 1997, for example). Here, since a high voltage of 40 V is applied to effect avalanche multiplication, it is difficult to make a pixel finer because of a crosstalk problem and the like. In the case of this sensor, the pixel size is 11.5 μm×13.5 μm.
Moreover, another avalanche multiplication type image sensor has been disclosed (refer to IEEE J. Solid-State Circuits, 40, p. 1847, 2005, for example). Regarding this avalanche multiplication type image sensor, it is desired to apply a voltage of 25.5 V to effect avalanche multiplication. Therefore, in order to avoid crosstalk, a wide guard-ring layer or the like is necessary, and a large pixel size of 58 μm×58 μm is desired.
In this regard, not only the above-described problem in that a high drive voltage is desired to effect avalanche multiplication for enhancing the sensitivity, but also a photon shot noise problem occurs at the same time because of a reduction in the number of photons. That is, since the photon is a Bose particle, overlapping of particles occurs, and in the continuous light, there are parts, in which photons are dense, and parts, in which photons are sparse (photon bunching effect). The noise Nn due to this fluctuation is the square root of the number of photons Ns as represented by the following equation.Nn=√Ns 
Therefore, the SN ratio becomes Ns/Nn (=√NS), and as the number of photons Ns decreases, the SN ratio decreases at the same time.
This refers to that a proportion of the photon shot noise relative to the signal increases.
In such a case, not only the signal, but also the photon shot noise is amplified at the same time through avalanche multiplication. Consequently, if multiplication is effected while the proportion of the photon shot noise is large, that is, the SN ratio is low, the noise becomes large relatively and the image quality deteriorates significantly.